Can I run this LLM?
Can a AMD Instinct MI300X (192GB) run Gemma 4 26B-A4B (MoE)?
Gemma · Apache 2.0 · estimated for a 192.0 GB machine at 4K context.
Yes — it runs great.
Best fit: Gemma 4 26B-A4B (MoE) at FP16 · 50.5 GB · 206–310 tok/s on a AMD Instinct MI300X (192GB).
Every quantization on this machine
| Quant | Size | Speed | Fit |
|---|---|---|---|
| Q3_K_M | 12.7 GB | 302–452 tok/s | Runs great |
| Q4_K_M | 17.0 GB | 287–430 tok/s | Runs great |
| Q5_K_M | 21.2 GB | 274–410 tok/s | Runs great |
| Q6_K | 23.2 GB | 268–401 tok/s | Runs great |
| Q8_0 | 26.9 GB | 257–386 tok/s | Runs great |
| FP16 | 50.5 GB | 206–310 tok/s | Runs great |
Open the full calculator →Everything a AMD Instinct MI300X (192GB) runsGemma 4 26B-A4B (MoE) on other machinesWhat to buy to run it →
Ran Gemma 4 26B-A4B (MoE) on a AMD Instinct MI300X (192GB)? Add your real numbers →